Forecasting Tehran's bourse price index using return-based fuzzy time series

During the recent years extensive researches have been done on fuzzy time series. In many of these studies, universe of discourse and relevant intervals have been determined based on levels of price or data; in this study a new type of universe of discourse is established based on rate of return concept in financial markets.
Another point that has a significant effect on the performance of fuzzy time series models is the length of intervals, therefore doing research in this area became an interesting topic for time series researchers, there are some studies on this issue but their results are not good enough. So we propose a novel simulated annealing heuristic algorithm that is used to promote the accuracy of forecasting. The experimental results show that proposed model (RBFTS) is more accurate than existing models on forecasting Alabama university enrollments data. At the final step, Tehran’s bourse price index (TEPIX) is used as a case study for forecasting. The obtained results indicate a good forecasting performance on this test problem.